The Evolution of Court ODR

So why don't you start out and just say a little bit about yourself. You're a professor at the University of Michigan Law School. How long have you been there? And a little background on your connection to ODR.

Great. Happy to do it. Yeah. So my name is JJ Prescott. I'm a law professor at the Michigan Law School. I also have an appointment in the economics department because I got here through both a JD and a PhD and my research really started off focusing on empirical work, particularly in the criminal justice system. And so I've been here since 2006, wrote a lot of different papers on criminal justice and settlement and litigation. Also more recently on non-compete in the last couple decades basically. But my interest in ODR came about in a slightly different way. I actually was spending time with a student who was fantastic, a really great student, but who wasn't interested in becoming a lawyer. And we started kicking around problems in the world. And I had been interested in access to justice, particularly in the criminal justice system space for a long time. And we started talking about warrants to appear for failure to appear and warrants for failure to pay. We focused on the fact that it's so interesting that so much of what that is about is the fear that people have about going to a courthouse when they don't know what to expect and no real way to solve the problem in any other way. So we imagined some way of allowing people to surrender safely to the court so that they could get on some sort of payment plan or reschedule a hearing without taking even what is oftentimes a very small risk of being jailed for having that warrant outstanding. So that was our initial idea, and that got us talking to courts and starting to think about the logic of how something like that would have to work, what would be important, what wouldn't be important, what would the barriers be. And that's where it all started.

Wow, amazing. And I love when I teach at law schools when I find someone who has decided over the course of their legal education, "You know what? I don't know if I wanna be a lawyer. I don't know if I want to take the bar exam." But, they're smart and they've learned a lot about justice and I zero in on those folks 'cause I want to recruit them to the ODR field. So I love that it was a student who got you initially thinking about this. So talk a little bit about how that turned into Matterhorn and the work you did at Matterhorn.

Sure. So it started with me and Ben Gubernick thinking about how we were going to do this. Neither of us are technologists or developers. At least at that time we weren't. And so we were basically two law guys, one of whom knows some about programming and statistical analysis and loves to work with data. But that wasn't really enough. So we spent some time at the University of Michigan in their office of Technology transfer, trying to identify someone who could help us connect some of the dots, think about how we could get the right skills to maybe build something. We wound up connecting with MJ Cartwright. We had an incredible meeting. She had worked in the healthcare space for a long time also on remote tools to help people.

Oh yeah. And MJ is a force of nature. Absolutely.

Yeah. So I still remember we were at Savas in Ann Arbor talking about this, and it was really something. She'd really never heard about any of the things we were talking about, warrants for people who were too poor, didn't know how to resolve them, and how there were millions of these things out there and they caused people to not call the police when they need help and all sorts of problems. And with her help, we were able to connect with a software development shop locally and start to build out something that could be piloted. And then I went into work, reaching out to nearby courts, talking with them about this warrant resolution idea, and pretty quickly they said, "that sounds great." We see the problem, what makes our courts really difficult for people? And also for us, our traffic disputes, we have a ton of 'em. They take a ton of time. There are people who come here and wait for five hours to have a short conversation to negotiate their ticket, if their record is a good one, for example. And interestingly, right after that, I got a ticket. In fact, for a long time in my early presentations, I had a deck of slides of my daughter in the back of the car, 'cause that's where she was. And the police officer said, "Hey, listen, your record is clean. It's great. If you go downtown, go down to the courthouse and make an appointment, they'll surely bump you down to something that won't put points on your record." And I said, "If that's true, why don't we do that right here?" He said that's not really something I can do. So that's what I did. I made an appointment. I went down, waited about four hours, and there were a hundred other people doing exactly the same thing. I went in, I had a conversation with the prosecutor. He opened up this paper file with two pieces of paper in it, and he said, "Mr. Prescott, I can offer you impeding traffic." Because it's the first time you've done something like this and I said, I'll take it. And he said, okay, we're done here. And I just, it blew my mind that all these people, hundreds and hundreds of hours of waste for these little minor conversations. And that's when I knew, listen, I care about warrants. I care about, especially people who are poor, who are scared of the system and can't use it. But even people who get traffic tickets and who can pay them but don't wanna, don't want to just write the ticket, they want to either contest it or talk to somebody about it. They're really also pushed out of courts, Umhmm, because of these, if you have a $150 ticket and you get paid by the hour, it basically costs you more to take a day off work to deal with a ticket than to just pay it. And in that case, absolutely, you really don't have a right. You really don't have a right to challenge a ticket in that situation. You just don't, it's just easier for you to accept whatever the government's allegation is.

And in the literature, I think Marco Lanter was the one who first talked about lumping it, which is just giving up. And that's probably the number one dispute resolution approach across the United States, which is I look at the cost benefit analysis of me pursuing redress, and I just say and as you say, the opportunity cost for me to take time off and go down and sit in a courthouse and wait for that conversation is higher than the cost of the ticket. So I might as well just pay it.

And we see that in small claims court too, Colin, where it's pretty normal to have 50, 60, 70% default rates. And some of that is because people don't know they're being sued, but a lot of it is let's say I have a good argument and I got a 50-50 chance of winning. It's gonna take me a lot of time to get down there and, so even if I win, I lose. And if I lose by default, they still have to come and find me. And it's actually pretty, pretty close. It's a close call. Of course, deciding not to show up has long-term consequences that aren't that aren't great for people. But a lot of people don't have the luxury of putting off dealing with things right now.

And I want to compliment you because I think Matterhorn ended up building what I think is actually the most successful court ODR program in the United States around these traffic cases in Michigan. And I think you had 30, 40 courts that signed up to use the platform over time and it's been running for 10 years and the courts all love that program. So we're actually coming in and trying to help out some of those courts too, with a next generation of it. But, it was your vision, JJ and all the great work that you and MJ did going around and building these relationships, because a lot of these courts that, trust me, I've done this, I've gone into courts and said, "Hey, you should use technology to do this more efficiently." and they look at you like, "Eh, we've tried to do stuff like this in the past and it hasn't worked out," but to actually build it, launch it, and then have it work successfully for, how many cases are we talking? Probably hundreds of thousands of cases.

Hundreds of thousands at this point. Yeah.

It's amazing. And thinking about your personal experience of going down and Thank you. I'll praise Colin. 'cause why not? But to be honest, I think a lot of the progress was a result of me being a law professor and me coming from the University of Michigan because those two things together within the state gave me a certain amount of credibility. I do think that you have a lot of new technology companies that have solutions, and I think people are rightly skeptical about whether or not they're as good as they say they are. But when a technology solution is backed by somebody who is also a lawyer and who understands what it is that judges do, I think that makes a pretty big difference. And honestly, I'm sure you've heard this a million times too, a court responding that seems so easy. We would obviously be doing it if it were that easy, so it must not be allowed. There must be some problem that you're not talking about.

And it's very interesting that you say that because now that I'm doing some work in Michigan, and I'm backed up, my boss is Bridgette McCormick, who was also a law professor at Michigan. And obviously went to the Supreme Court, but I think that there's a great book that I love by a guy named Simon Sinek and he talks about, "Start with Why," and he said, when you're selling something, people don't buy from you because of what you're selling. They buy from you because of why you're selling it. And the big brands are, they all have a very strong why. And I think that look, as a vendor who is selling to courts, I understand the skepticism because there are a lot of companies that go out there and promise the moon and the stars. And then when you get into it, you're like, oh man, this is not what I thought it was gonna be. But if you're buying from somebody and you know their value. And I think that's true with you as a law professor. I think that's true with Bridget. I like to think it's true with me because, my younger kid's name is Alexander David Rule for the initials, and he's 25. Like... people know the why. They believe. It's okay, I understand why you're doing this and I don't understand all of the what, but we can figure that out together so long as I have the why squared away and I think that you're right, I think a lot of people did trust you and Matterhorn and MJ because of the roots of where the project came from and because also, you were in Michigan, I think those were all really important things that got the court to say, "I can trust these guys and let's dive into the unknown a little bit with them.

And as once you have a few courts, especially courts with leaders that are known to be leaders, once they dive in and the world doesn't end others are easier to persuade. Especially once they start talking up your solution, explaining why it's making their lives easier, how much time they're saving, and so on. And that was also the case with Matterhorn. You always have problems getting things going in the beginning, but pretty much, to a judge, to a user, people were always super happy, even the ones who were very skeptical.

Sure. So let's talk about where you've gone since then. You shared with me a really great article that you did for the Alaska Law Review, talking about court connected ODR, but what are your research interests now? What are you focused on?

Yeah, so I write in other areas as well, but in the ODR space, I'm interested in AI. My training as an empirical economist makes me really interested in machine learning more generally. And I have for a long time been interested in the use of analytics in ODR. And also in moving on to adjudication, online adjudication that might even happen in an asynchronous, text-based way. And so this article in Alaska came out of a symposium put on by the Alaska Law Review, interested in access to justice and the use of technology. My contribution related to the use of technology and, in early... I guess at the end of 2023, they had launched a new system that had a number of features that I really liked. It had limitations as well, of course, and I talk about those in the article. But some of the nice features of it I thought were ones that were worth talking about and sharing. And some of those were, it was really easy to sign up and you didn't have to have an existing case. So a lot of times, there has to be an existing case filed in a court so that there's some record of the legal matter. And then you can maybe access that's not true in Alaska. When A-K-O-D-R launched, they decided to make it really so that you could start a claim right from your phone. Actually not from your phone. That was one of the challenges, but from your computer you could list somebody and invite them to a negotiation space. And honestly, it wasn't even restricted to an Alaska resident. Of course, if you wanted to turn it into an enforceable order later on, I assume that would require something more than just signing up online from another state. But I was able to, for example, ask one of my RAs to make a claim against me in small claims. And we were able to use the system to negotiate this fake claim to see how it would work. And so it was so easy to get on and start using that I thought, okay, this is a great way for a state that is relatively small in terms of population to really potentially bring in testers, users to use this. And then on top of that they had plans and I hope still have plans in the future to move on to online adjudication asynchronously. So they had used it for negotiation and then it also included mediation where the courts would provide a mediator who would enter into an asynchronous mediation and help both sides. And the next step on this process was to move on to actually having a decision maker who would be making determinations as a result of of the exchanges going on. And, they were working out some of the details when I last talked to them about this. Like how would the back and forth go, what sort of procedural requirements would exist in this space? What kind of finality would a decision have? Those are all really hard and interesting questions, but Alaska was pushing—as far as I know, maybe you know of another place—farther than anybody else I knew at the time.

No, you're a hundred percent right. I think Alaska has been very innovative in thinking about all of these tools and the ways to do it. One of the frustrations I had when I first started working with Tyler on Modia, my prior ODR platform, was that first of all, they said we have to have the case e-filed and a response filed before we start ODR. And that frustrated me because obviously there's filing fees associated with that. And also it imposes a certain narrative on the dispute, because going through the e-filing process says this is me versus you. And again I'm on record as saying, initially, when I started working in ODR, I never wanted to sell ODR to the courts. I wanted to get bulletin boards outside of all the courts that said, don't go in there, come to odr.com and we'll take care of you. And I think I love the idea that anybody can come and start using the platform. I think that that really lowers a lot of the barriers to entry in the use of these platforms.

And to take it a little further, they do require, when you make a claim, that you state the nature of your legal claim against the person. And in the paper I ask, why do that? If I have some sort of dispute with my neighbor or somebody in my neighborhood, I may not really even want to characterize it as a claim or a dispute or even put a label on it. Why not allow something that essentially lets people start a conversation. You don't really need a case type until it moves into the courts.

And oftentimes the types that the courts have aren't meaningful or helpful to the parties. Saying, I have a small claims dispute. Really all, the only thing that defines that is the monetary value of the case. There could be a thousand different types. It could be a barking dog or an IP case, or a property case or a consumer case. But just to say this is a small claims case. Is that helping the parties understand the path to resolution?

Totally. Yeah. And there's one other feature that I thought reduced the potential utility of A-K-O-D-R, which was the fact that only somebody who was in the role of a plaintiff could initiate the ODR. So if you had a claim and you had made a claim, you could start it, or if it was... I think the definition that they use is: the person who is claiming they are owed something or that they deserve something, they have to start ODR. Which I thought - Listen, some of the people are gonna benefit most from ODR and are probably gonna put the time into are gonna be defendants in debt cases, for example, who live too far away. They're the ones who might say, okay, you filed a case against me, or you made a claim against me. I would like to use this system because it will make my life a lot easier. But in Alaska, in its initial implementation, that was not possible. So you could only initiate ODR aggressively as a plaintiff. You couldn't use it as a way to reduce the costs for somebody who was defending.

Another frustration I had with the way that Tyler framed ODR, and this was, this was early, this was 10 years ago, was that it was always the facilitative piece. It was let's focus on the negotiation. Let's focus on the mediation. And I think, again, I think you cite in your article, the National Center for State Court's definition of ODR, which is an online collaborative workspace where the parties can convene to resolve their dispute or case, which I think is a pretty good description of part of ODR. But obviously, and I wanna go back to what you said before, there is an evaluative piece. I'm building a lot of online arbitration systems where it goes all the way through to an award being made by an arbitrator or a panel of arbitrators. And it's interesting to me that court ODR has really steered clear of that. And I think the reason why Tyler felt that way was because they said that's what the courts do. The judges are in the business of evaluating cases and rendering decisions. Let's build ODR to just focus on this facilitative phase in advance of the hearing. And if the parties can work it out, great, you can cancel the hearing. But in the United States, I don't think there's been any examples of court ODR that goes all the way through the evaluative process and rendering a decision. Now other courts around the world have embraced that, even in British Columbia, which isn't that far away from Alaska, the Civil Resolution Tribunal does have an all online process where you never have to show up in person and you can go all the way through the decision. And if you get a decision from the Civil Resolution Tribunal, it's just as enforceable as if you went through the face-to-face process. And it's worked great. They have tons and tons of data that they pulled out of the CRT showing that parties really appreciate that all-online option. So I really appreciate you calling this out because I do think there's huge potential for evaluative ODR in the courts and I know Alaska is far ahead on thinking about this, but I think it's coming for other states as well,

and it doesn't have to be a judge, to the extent judges are worried about it. You're exactly right that you can have other parties serving in this role. Arbitrators. If you think about the types of evaluative in-person procedural tools that are out there, case evaluation in Michigan, for example, where you have three experts sit on an abbreviated presentation of the facts. They announce a value on the case and then the parties use that and creates an offer of judgment type of situation. It doesn't resolve the case, but it certainly changes the case in an important way. And of course, if you just use ADR and have somebody and everybody agrees to go with what the arbitrator says, there's no reason, I don't think that maybe the next step for courts ought to be to do that. To allow the parties to say, even if judges don't want to play that role, to have somebody or to have a stock of arbitrators who can do this sort of work.

Absolutely. And I mean we're now part of the American Arbitration Association, so I'm learning a lot about good arbitration, how do you design these processes so there's transparent and they're, they're just, and I think there is a lot that can be learned from some of the arbitration mechanism and vice versa. I think courts can teach things to arbitration institutions and arbitration institutions have lessons as well. One of the things that's interesting is that Matterhorn, in Arizona for a while, was deployed in a pilot in some criminal misdemeanor spaces.

And there's a lot, I think more questions once you have an evaluative part of this about, public access to the proceeding.

Sure.

And if you decide to opt out of the system and you're using arbitration, fine, that's basically akin to settlement. You're resolving it privately. But if you are going to have an online hearing, essentially in asynchronous text, how exactly do we make sure that the public is able to access that? To me, these are really interesting questions, and of course I can imagine a few different ways that might be, but I think an important research question is how will people react to them?

Absolutely. So, some of my interests are, of course, both in improving the efficiency of courts—how do we do that?—and also in how people's views about courts, judges' decision-making, and lawyers' behavior are going to change in ways that we should anticipate, so that we can either avoid pitfalls or improve their behavior.

Well and I can see, I can, immediately arguments pop to mind as to how these online asynchronous processes can actually improve transparency. Totally. In terms of people knowing what's going on in these cases. When I first came to Tyler… I'm a non-lawyer mediator. My background's in public policy and technology. So I said, look, I don't do anything on the criminal side. Like this is really about civil cases and just, working out issues. But I had an opportunity to meet a lot of judges on the criminal side who were working with Tyler, and they said, you realize most criminal cases are resolved through negotiation. And I said, what do you mean? And they said it's the public defender and it's the district attorney and they're making offers back and forth, and trying to figure out what an appropriate resolution is. So again, it's not the same thing as putting the victim and the offender together in an environment to negotiate a resolution. But clearly these tools are relevant on that side of the equation as well. And I do think that the kind of historic bias towards face-to-face is going, the culture is changing around the legitimacy of these online processes. And, we're starting to realize that, requiring face-to-face participation is its own form of privilege, especially in a place like Alaska where you may have one side is right across the street from the courthouse and the other one's three hours away. So by insisting that all happened face to face you are creating a power imbalance between...

Three hours and a plane flight. Many communities there are not even connected by roads to...

Exactly. Exactly. I, this is such a fascinating conversation, JJ and I want to continue it with you, but let me take it back to something you said, and this is squarely in your expertise area, which is about data, which is about research. And one of the things that we didn't have when we first started doing the court ODR stuff back in 2012, 2013. We didn't have any data about it. So a lot of this was promise, hey, maybe if we do this, we will have these benefits. But I think one of the things that was great about Pew and the money they put into court, ODR, was they funded researchers who don't have a dog in the fight to take longitudinal looks at some of these court ODR programs and then see, what works the way that we expected it to work and what didn't work so well. And I think your quantitative background is really valuable. I know you're constantly asking yourself, okay, what data do we need here to know what's really going on? But as we move forward on the next generation of court ODR, what data would you like to get your hands on to really make sure that we're headed in the right direction?

That's a really good question. I, when I look at courts the first thing that I find really frustrating is just how siloed they are in terms of their data resources. There's so much that is essentially replicated in courts that might be easily gained from other government agencies. You have a whole hearing on, for example, ability to pay, but that question has already been answered by other departments within the government, and yet you don't have it. So for me, I think both in terms of thinking about outcomes and also in terms of thinking about efficient sources of data to courts outside of what happens in the court would be really useful. But in terms of thinking about what other, in terms of rigorous research, one of the biggest challenges is making sure that people who use these tools, the kinds of information we're collecting from them and how we're collecting it from them is the same as the kind of data we're collecting from people who don't use the courts at all, or who use the courts in more traditional ways. And that's just a real challenge. I've tried to overcome it in a few different ways. So we give a survey to people who've used ODR and ask them some questions, but now we have to survey people who have used in-person proceedings and how exactly do we do that? If we send a survey to somebody three days after their ODR outcome, so they have a little bit of time to think about it and they fill it out, how would I get the equivalent from somebody who is walking out of a courthouse. Do I ask them to give me their email address so I can send them a survey in three days. And really, I don't have a great, I don't have a great solution to any of that other than encouraging courts to build the data infrastructure and connections to their users so that subsequent contact with them is a little bit easier. And this is of course a great source of problems for courts because people, especially people who have misdemeanors or who have outstanding warrants for failure to appear or failure to pay, they move around a lot. You have a traditional mailing address for them that oftentimes becomes stale very quickly, whereas a phone number won't become stale. And yet, a lot of courts don't really collect phone numbers. So I don't know if that's a great answer. I don't have a great solution, but I'm always worried about control and treatment groups and trying to make sure that I can really compare apples to apples, right? Instead of just the people that I can connect to in those groups who wind up being differently selected depending on how you try to reach them.

Yeah. It's interesting I've wrestled with this over the 25 some odd years that I've been in ODR. I can get feedback from my participants at the end of the process and say, yeah, what'd you think about the process? But one of the challenges I've had is trying to separate outcome from perceived procedural justice. Yeah. Because if you win, you like it, if you lose, you didn't like it. And I think that there are, I think about Jared Kern at MIT, he's got this thing called the subjective value indicator, which is a series of questions that he's tested that gets to more of the perceived procedural legitimacy of the resolution as opposed to the outcome. But one of the things, and so I have a lot of data that I gather, and I know Matterhorn did this too, at the end of the process, but what confounds me is the longitudinal data. Because I have courts say to me how do I know these outcomes are durable? Yeah. If I do a face-to-face process, maybe I don't see 'em for 10 years, but if they do an online process and it's really easy, maybe they're back in a year and then they wanna renegotiate their terms and I can't do that. Right. As a provider, I don't have the ability to follow up with parties a year or two or three down the road. And so much like you say, the access question of people who are in these cases, they move around, it's hard to follow up with them. That longitudinal data I think is really important. And the only way that we're gonna get rapid adoption of these techniques is if we can prove with data that not only that they're matching the quality of face-to-face outcomes, but even exceeding in a sense, in terms of, key variables like access and procedural satisfaction.

Yeah you can imagine trying to essentially match people who. Randomly, let's say we have a situation where we think there's some exogenous variation that allows some people to use ODR and some people just can't maybe by time of adoption or something else. And then you look at court data down the road, maybe at the state level, so you feel like you're seeing other sorts of things. If you're worried about people coming back. Those are easier to observe and people appearing either because they have subsequent misdemeanors, something we might call recidivism or just they are having new touches with the courts later on. It would be interesting and I think feasible to try to do that. How people feel a year later. That seems to me, what I thought early on, I don't know if you felt this way when Matterhorn first started, people would search for their case. They'd resolved their case, they'd put in their information. Then the case would disappear and there was nothing. It was the court's data and there was no long term account that somebody would have like a justice account with the court. But from the beginning I thought this seems silly. If people, the court has long-term relationships, and it really does because it holds records of people. It oftentimes knows about their obligations. It's oftentimes looking for them, but having an ongoing account with people that would allow and maybe what it winds up being driven by is something at the state level, something like your the same way you'd pay your state income taxes or some other relationship that would allow courts to be able to stay connected with their users over time, and that would facilitate a... I would like to think surveying people down the road and I think you're right that, going into a court, it can be both incredibly frustrating, but it's also more memorable. So it might turn out that people who use these online tools, they don't even remember that they went to court. And I can see that being a good thing, but I can also see it being a bad thing when we want people to understand that these resources are out there and remember how to use them well.

Yeah, part of the goal of a good resolution process is enabling the parties to put it in their rear view mirror. And just not think about it anymore. Okay. We got a resolution. It's fair. I move on. I do my life. This is one of the things we've seen on the family side. When we built the Rect advisor, parties would come in, go through the divorce and separation process, and they'd come up with an agreement and then oftentimes mediators and courts would say, okay, we got the agreement. Thanks, bye. But then later they would need to modify their understanding around child support. Obviously, the costs of a 5-year-old are different than the costs of a 15-year-old. But, then they would come back and they'd kinda have to start again. Yeah. And we were saying, why, how does that make any sense? Once we have the agreement, we should put it into an online platform and if they need to make an adjustment, they should be able to log in and access it again and then say, hey, can we adjust it this way? And the other side? Yes, that's fine. And then you update it. And if you look at platforms like Our Family Wizard and some of these co-parenting management platforms, that's what they're doing. They're filling in for the courts because it's about the ongoing adherence and evolution of the agreement as opposed to, okay, we got an agreement. Both sides signed it, we're done. Good luck to you. Call us if you have a problem.

And the agreement is an active agreement being essentially supervised by the court. So the idea that there's like an ongoing relationship with the court makes sense and it would be nice if we could make it so it's not something you had to think about every single day. But you also want to be able to access it easily and be able to keep records and figure out what had happened in the past easily as well. And I just think about using the bank software. I can go on when I need to use a tool that my bank offers and I can find my history, I can find out when I did things I can, it really helps me to understand...

Amazon will tell you what you bought 10 years ago. Yeah, sure. That makes sense. It makes sense. And it, and now and the retention and access to those records is really easy. Again, with technology, whereas it might be a lot harder for a court to have somebody come in and say, Hey, I want to pull my file. Oh wait, this is 10 years ago. Where did we put that? Yeah. So look, JJ, I could talk to you all day about this stuff, and I hope we can do a part two because as we make progress, I definitely want to benefit from your wisdom and eager to partner with you on the academic side too 'cause you're doing such pathbreaking work. But I really do appreciate all your contributions to the ODR field. And thank you for spending the time today and I look forward to continuing the discussion with you.

Absolutely. My pleasure, Colin. Thank you.

The Court ODR podcast is brought to you by odr.com. To learn more about odr.com products and services, please visit us at odr.com.